A H+/organic cation antiporter (multidrug and toxin extrusion 1: MATE1/SLC47A1) plays important roles in the tubular secretion of various clinically important cationic drugs such as cimetidine. We have recently found that the regulation of this transporter greatly affects the pharmacokinetic properties of cationic drugs in vivo. No information is available about the regulatory mechanisms for the MATE1 gene. In the present study, therefore, we examined the gene regulation of human (h) and rat (r) MATE1, focusing on basal expression. A deletion analysis suggested that the regions spanning −65/−25 and −146/−38 were essential for the basal transcriptional activity of the hMATE1 and rMATE1 promoter, respectively, and that both regions contained putative Sp1-binding sites. Functional involvement of Sp1 was confirmed by Sp1 overexpression, a mutational analysis of Sp1-binding sites, mithramycin A treatment, and an electrophoretic mobility shift assay. Furthermore, we found a single nucleotide polymorphism (SNP) in the promoter region of hMATE1 (G−32A), which belongs to a Sp1-binding site. The allelic frequency of this rSNP was 3.7%, and Sp1-binding and promoter activity were significantly decreased. This is the first study to clarify the transcriptional mechanisms of the MATE1 gene and to identify a SNP affecting the promoter activity of hMATE1.
- tubular secretion
- basal transcriptional activity
- single nucleotide polymorphism
- multidrug and toxin extrusion 1
renal elimination, including glomerular filtration and tubular secretion, is a major clearance mechanism for many drugs. The renal tubular secretion of drugs takes place primarily in the proximal tubules as an interplay of drug transporters located in the basolateral and brush-border membranes, respectively. For example, organic cations are taken up by membrane potential-dependent organic cation transporters and then extruded into the lumen by H+/organic cation antiporters (14). Molecular cloning of membrane potential-dependent organic cation transporter 1 (OCT1) was reported in 1994 (12), and at present, three kinds of organic cation transporters (OCT1–3) from various species have been identified and characterized (7, 14, 36). In contrast to OCTs, attempts at the molecular identification of apical H+/organic cation antiporters have failed for more than a decade.
Recently, mammalian orthologs of the multidrug and toxin extrusion (MATE) family, which confers multidrug resistance on bacteria, have been identified in humans (19, 25), mice (25), rats (23, 34), and rabbits (37). Two functional isoforms, MATE1/SLC47A1 (human, mouse, rat, and rabbit) and MATE2-K/SLC47A2 (human and rabbit), have been characterized in terms of tissue distribution, membrane localization, and transport characteristic and have been demonstrated to work as H+/organic cation antiporters (2, 19, 20, 23, 25, 33–35, 37). MATE1 is able to transport various organic cations such as tetraethylammonium, cimetidine, and metformin and the zwitterion cephalexin using an oppositely directed H+ gradient (33–35). Human (h) MATE1 is primarily expressed in the kidney and liver and is located at the membranes of the luminal side (19, 25). Rat (r) MATE1 is also present in the brush-border membranes of renal proximal tubules (20), and the intrarenal expression of rMATE1 is limited to the proximal tubules (34). Our group (20) also demonstrated that the expression level of rMATE1 was remarkably reduced in 5/6 nephrectomized rats and that the level of rMATE1, but not of rOCT2, correlated well with the tubular secretion of cimetidine in female rats (r = 0.74). Furthermore, very recently, our group identified cysteine and histidine residues essential to the function of rat and human MATE1 (2). Hence, functional and expressional aspects of MATE1 have been clarified in much detail, but there has been no information about the transcriptional regulation including basal promoter activity of this transporter.
The core promoter and proximal promoter regions contain elements that control the initiation of transcription, and therefore, essential regions that harbor functionally relevant polymorphisms may have significant effects on gene expression (6). When such polymorphisms occur in the promoter region of pharmacokinetic-related genes, drug disposition, efficacy, and toxicity can be influenced. For example, a genetic polymorphism of the uridine diphosphate-glucuronosyltransferase 1A1 gene, UGT1A1*28, a variant sequence in the promoter region, contributed to individual variation for adverse events in patients administered the anti-cancer agent irinotecan (13). Thus a search for single nucleotide polymorphisms in the promoter region (rSNP) of drug transporters may provide useful information on interindividual differences in drug disposition.
In the present study, therefore, we characterized the promoter activity of hMATE1 and rMATE1 to identify the minimal region and cis-regulatory elements required for the basal expression of MATE1. Furthermore, we performed rSNP analyses for hMATE1 and found that SNP at G−32A, which belongs to a Sp1-binding site, affected the promoter activity of the gene.
MATERIALS AND METHODS
[γ-32P]ATP was obtained from GE Healthcare (Little Chalfont, UK). Restriction enzymes were obtained from New England BioLabs (Beverly, MA). Antibody for Sp1 (H-225) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mithramycin A was purchased from Sigma-Aldrich (St. Louis, MO). The plasmid CMV-Sp1 was kindly provided by Dr. Robert Tjian (University of California, Berkeley, CA). All other chemicals used were of the highest purity available.
Determination of putative transcriptional start site.
To identify the transcription start site of hMATE1 and rMATE1, we carried out 5′-rapid amplification of cDNA ends (5′-RACE) using human and rat kidney Marathon-Ready cDNA (Clontech, Mountain View, CA), respectively, according to the manufacturer's instructions. The primers for 5′-RACE were as follows: a gene-specific primer for hMATE1 (GenBank accession no. FLJ10847), 5′-CAGGACCAAGAGCGCCCGCAGCTCTTCT-3′ (209/182); a nested gene-specific primer for hMATE1, 5′-CTGGCGCGGGCTCCTCAGGAGCTTCCAT-3′ (114/87); a gene-specific primer for rMATE1 (GenBank accession no. NM_001014118), 5′-GAGCTGGGCCAAGAACGCAGGACCC-3′ (170/146); and a nested gene-specific primer for rMATE1, 5′-CTGGTCCCGGCGCAGGCTCCTCCAAGA-3′ (57/31). The polymerase chain reaction (PCR) products were subcloned into the pGEM-T Easy Vector (Promega, Madison, WI). All PCR products were sequenced using a multicapillary DNA sequencer RISA384 system (Shimadzu, Kyoto, Japan).
Construction of reporter plasmids for hMATE1 and rMATE1 promoters.
The human or rat promoter region upstream of the transcription start site was amplified by PCR using human genomic DNA (Promega) or rat genomic DNA (Clontech) with the primers listed in Table 1. The PCR products were subcloned into the firefly luciferase reporter vector pGL3-Basic (Promega). The full-length reporter plasmids are hereafter referred to as hMATE1 (−2408/+13) and rMATE1 (−2422/+24). The 5′-deleted constructs were generated by digestion with appropriate restriction enzymes. The hMATE1 (−65/+13) and rMATE1 (−38/+24) constructs were generated by PCR with the primers listed in Table 1. The site-directed mutations in putative Sp1-binding sites were introduced into the hMATE1 (−65/+13) construct with a QuickChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA) with the primers listed in Table 1. All PCR products and deletion constructs for reporter assays or rSNP analyses were sequenced using a multicapillary DNA sequencer RISA384 system (Shimadzu).
Cell culture, transfection, and reporter gene assays.
LLC-PK1 and HEK293 cells were obtained from the American Type Culture Collection (ATCC CRL-1392 and -1573). Cell culture, transfection, and reporter gene assays were carried out as described previously (1, 22, 29).
Western blot analysis.
Preparation of nuclear extracts from LLC-PK1 cells was described previously (29). The nuclear extracts of LLC-PK1 cells (125 ng) were solubilized in SDS sample buffer, separated on a 10% polyacrylamide gel, and transferred onto polyvinylidene difluoride membranes (Immobilon-P; Millipore, Bedford, MA) by semidry electroblotting. Blots were blocked with 5% nonfat dry milk in Tris-buffered saline (TBS; 20 mM Tris, 137 mM NaCl, pH 7.5) with 0.1% Tween 20 (TBS-T) for 1 h at room temperature. Blots were washed in TBS-T and then incubated with the anti-Sp1 polyclonal antibody (1 μg/ml, overnight at 4°C). The bound antibody was detected on X-ray film by enhanced chemiluminescence with a horseradish peroxidase-conjugated anti-rabbit IgG antibody (GE Healthcare).
Recombinant human Sp1 (rhSp1) and recombinant human Sp3 (rhSp3) were obtained from Promega and Alexis (Lausen, Switzerland), respectively. The probes shown in Table 1 were prepared by annealing complementary sense and antisense oligonucleotides, followed by end labeling with [γ-32P]ATP using T4 polynucleotide kinase (Takara Bio, Otsu, Japan) and purification through a Sephadex G-25 column (GE Healthcare). The composition of the binding mixture containing rhSp1 was based on Martín et al. (17), and compositions of rhSp3 binding buffer were as follows: 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 10 mM Tris·HCl (pH 7.5), and 4% glycerol. Labeled probes were added to the binding mixture and incubated for 0.5–1 h at room temperature. The DNA-protein complexes were then separated on a 4% polyacrylamide gel at room temperature in 0.5× Tris-borate-EDTA buffer. The gel was dried and exposed to X-ray film for autoradiography.
Genomic DNA was extracted from peripheral blood from 88 patients with renal diseases (47 men and 41 women; range 14–78 yr) and normal parts of the kidney from 21 renal cell carcinoma patients (21 men; range 39–77 yr) using a Wizard Genomic DNA purification kit (Promega). About 120 bp of the promoter region of the hMATE1 gene were amplified by PCR using the forward primer 5′-CGCAGTGGTGCAGAGAGAGGTGCAA-3′ (−154/−130) and the reverse primer 5′-AGTCACCCGCGGAGGCAGAAATCAC-3′ (297/273). PCR conditions were as follows: denaturation at 94°C for 30 s, annealing and synthesis at 68°C for 3 min, for 35 cycles. All PCR products were sequenced using a multicapillary DNA sequencer RISA384 system (Shimadzu).
This study was conducted in accordance with the Declaration of Helsinki and its amendments and was approved by the Kyoto University Graduate School and Faculty of Medicine Ethics Committee.
The results are mainly expressed relative to pGL3-Basic and represent means ± SE of three replicates. Two or three experiments were conducted, and representative results are shown. Data were analyzed statistically with one-way ANOVA followed by Dunnett's test.
Determination of the transcription start site of hMATE1 and rMATE1.
Sequencing of the longest RACE product showed that the terminal position of hMATE1 and rMATE1 cDNA was located 22 and 20 nucleotides upstream of the start codon, which was 64 and 3 bp downstream of the 5′-end of hMATE1 and rMATE1 cDNA, respectively, registered in the National Center for Biotechnology Information database [GenBank accession no. FLJ10847 (hMATE1) and NM_001014118 (rMATE1)]. Therefore, the 5′-end of hMATE1 and rMATE1 cDNA in the database was numbered with +1 as the transcription start site in this study.
Determination of the minimal hMATE1 and rMATE1 promoter.
To determine the minimal region required for basal activity of the promoter, a series of hMATE1 (Fig. 1A) or rMATE1 (Fig. 1B) deletion constructs were transfected into LLC-PK1 cells, and luciferase activity was measured. In LLC-PK1 cells, the H+/organic cation antiport system is expressed in the apical membrane (9, 15, 28), and we hypothesized that transcription factors and/or cofactors required for the expression of MATE1 exist intrinsically in these cells. The longest reporter constructs of hMATE1 and rMATE1 showed an approximately three- to fivefold increase in luciferase activity compared with pGL3-Basic. The luciferase activity of both constructs was completely abolished with the −25/+13 (hMATE1) and −38/+24 (rMATE1) deleted promoter constructs, respectively, suggesting that the region between −65 and −25 for hMATE1 or between −146 and −38 for rMATE1 was important for basal promoter activity. We then performed a computational sequence analysis of these regions using TFSEARCH (www.cbrc.jp/research/db/TFSEARCH.html). This analysis revealed that the region proximal to the transcription start site lacks canonical TATA or CCAAT boxes. Instead, two GC-rich sites were observed, suggesting a possible contribution of Sp1 to the basal promoter activity of MATE1 (Fig. 2, A and B).
Western blot analysis.
To confirm the expression of Sp1 in LLC-PK1 cells, we carried out Western blot analysis for Sp1. As shown in Fig. 2C, an immunoreactive protein with a molecular mass of 106 kDa corresponding to Sp1 was detected in LLC-PK1 cells.
Functional involvement of Sp1 in MATE1 promoter activity.
To evaluate the functional involvement of Sp1 in MATE1 promoter activity, we performed experiments for the inhibition and transactivation of Sp1. Mithramycin A is known to bind to the GC box and inhibit Sp1 binding (4, 27). As shown in Fig. 3, treatment with mithramycin A led to a significant decrease in the promoter activity in a dose-dependent manner for both reporter constructs. Next, we investigated the effect of Sp1 overexpression on the MATE1 promoter activity in HEK293 cells (Fig. 4). Transfection of the CMV-Sp1 expression vector in LLC-PK1 cells remarkably reduced the activity of the pGL3-Basic (control) via unknown mechanisms, and therefore, HEK293 cells were used for Sp1 overexpression experiments. The promoter activity of hMATE1 (−65/+13) or rMATE1 (−146/+24) showed a dose-dependent increase in the cotransfection of CMV-Sp1, providing direct evidence that Sp1 enhanced the activity.
Identification of Sp1-binding site of hMATE1 promoter.
In the proximal hMATE1 promoter region, there are two putative Sp1-binding sites located between −65 and −25, and these sites were designated as Sp-A (−64/−55) and Sp-B (−40/−31), respectively. To determine whether these sites were important for hMATE1 promoter activity, we introduced a mutation at each site of the hMATE1 (−65/+13) construct (mutation A and mutation B; Fig. 5A). As shown in Fig. 5B, both mutants remarkably reduced the luciferase activity compared with the wild type. These results suggest that both Sp-A and Sp-B are important for regulating the hMATE1 promoter activity. Finally, to confirm that Sp1 binds to Sp-A and Sp-B sites, we performed EMSAs with rhSp1 and wild-type or mutational probes of Sp-A and Sp-B (Fig. 5A). Both wild probes A and B formed DNA-protein complexes (Fig. 6A, lanes 3 and 7), whereas no complex was formed in the absence of rhSp1 or mutational probes A and B (Fig. 6A, lanes 1, 2, 4–6, and 8). These results suggest that Sp1 stimulates the basal promoter activity of hMATE1 via Sp-A and Sp-B regions. It was reported that Sp3 as well as Sp1 also recognizes classic Sp1-binding sites (5). We then further examined the interaction of Sp-A and Sp-B with rhSp3. As observed in rhSp1, specific DNA-protein complexes in both wild probes A and B were detected (Fig. 6B).
rSNP Analyses of hMATE1.
We then sequenced the promoter region (∼120 bp) of the MATE1 gene in 88 patients with renal diseases and 21 patients with renal cell carcinoma and found the G−32A SNP at the Sp-B site. The frequency of this rSNP is summarized in Table 2. To evaluate the effect of the G−32A SNP on the promoter activity of hMATE1, we carried out an EMSA and a reporter assay. Both wild and SNP probes formed DNA-protein complexes, although less complex was formed with the SNP probe (−32A) (Fig. 7, lanes 4 and 8) than with the wild probe (−32G) (Fig. 7, lanes 3 and 7). No complex was formed in the absence of rhSp1 (Fig. 7, lanes 1, 2, 5, and 6). Furthermore, as shown in Fig. 8, the SNP construct (−32A) remarkably reduced the luciferase activity compared with the wild type (−32G). These results indicate that the G−32A substitution downregulates the basal promoter activity of hMATE1 by weakening the binding of Sp1.
In the present study, we performed a functional promoter assay of human and rat MATE1 genes and obtained convincing evidence of the involvement of Sp1 in the regulation of the basal expression of these transporters. This conclusion is supported by results of experiments involving the inhibition of mithramycin A, the overexpression of Sp1, mutagenesis of the GC-rich region, and EMSAs using rhSp1. In the proximal promoter region of the human and rat MATE1 genes, two GC-rich sites exist instead of a TATA box, and these features are conserved among species (Fig. 9). In a TATA-less promoter, Sp1 binds to the GC-rich region to recruit TATA-binding protein (26) and to fix the transcription start site (3). Furthermore, it was reported that the promoter activity is enhanced if multiple Sp1-binding sites exist (18). These findings suggested that Sp1 plays a significant role as a basal transcription factor through these GC-rich sites in the MATE1 promoter.
Furthermore, we found a SNP at a Sp1-binding site (G−32A, belonging to the Sp-B site) of the hMATE1 promoter for the first time and demonstrated that this substitution affects hMATE1 promoter activity by disrupting the binding of Sp1. It is strongly suggested that this rSNP influences the mRNA level of MATE1.
To date, many large-scale screenings of SNPs of drug transporters (mainly focusing on SNPs in the coding region; cSNPs) have been carried out to identify genetic factors involved in the interindividual differences of pharmacokinetics. So far, clinical implications for cSNPs in the genes for organic anion transporting polypeptides (16) and OCT1 (31, 32) have been demonstrated, but significant results have not been obtained for other SLC-type drug transporters. In the present study, we found an rSNP of Sp1-binding sites affecting the MATE1 gene. This type of SNP may be involved in the interindividual differences of pharmacokinetics. There are reports that rSNPs of Sp1-binding sites of the resistin gene contribute to type 2 diabetes mellitus susceptibility (24) and that those of the collagen type I α1 gene contribute to low bone mass and vertebral fracture (11). Further studies of the relationship between gene polymorphisms of MATE1 and the pharmacokinetic properties of MATE1 substrate drugs may clarify the clinical implications of this SNP.
Human and rat MATE1 mRNA is abundantly expressed in the kidney (19, 34), whereas Sp1 is expressed ubiquitously. The present study revealed the contribution of Sp1 to the transcriptional regulation of MATE1, but the mechanism of this tissue-specific expression has not been clarified. In the case of intestinal H+-coupled peptide transporter 1, the basal promoter activity was also determined by Sp1, but its tissue specific expression was regulated by the intestinal specific transcription factor Cdx2 (29, 30). Currently, there is little information about the renal specific transcription factor. It has been reported that Sp3 as well as Sp1 also recognizes classic Sp1-binding sites, and there are a number of reports that the relative abundance of Sp1 and Sp3 should allow regulation of gene activities (5). We confirmed by EMSA that a recombinant Sp3 protein bound to the Sp-A and Sp-B sites of the hMATE1 promoter (Fig. 6B). Another possible mechanism is the methylation of the GC box in the promoter region. Methylation of the cytosine residue in the sequence of 5′-CpG-3′ is an epigenetic modification, and recent reports have revealed that this epigenetic modification contributes to the tissue-specific expression (8, 10, 21). Further studies are needed to clarify the tissue-specific expression of MATE1.
In conclusion, the present study clearly indicates that Sp1 functions as a basal transcriptional regulator of the human and rat MATE1 gene through the two GC boxes, and this system may be conserved among species. Furthermore, we have identified an rSNP of the hMATE1 gene (G−32A) (belonging to a Sp1-binding site) that affects the promoter activity.
This work was supported by the 21st Century Center of Excellence Program “Knowledge Information Infrastructure for Genome Science,” a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and a Grant-in-Aid for Research on Advanced Medical Technology from the Ministry of Health, Labor and Welfare of Japan.
We are grateful to Dr. Robert Tjian (University of California, Berkeley) for the Sp1 expression vector.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society